Zobrazeno 1 - 10
of 546
pro vyhledávání: '"Ahmed, Kareem A."'
Autoregressive models have demonstrated an unprecedented ability at modeling the intricacies of natural language. However, they continue to struggle with generating complex outputs that adhere to logical constraints. Sampling from a fully-independent
Externí odkaz:
http://arxiv.org/abs/2410.13111
Large Language Models (LLMs) are typically shipped with tokenizers that deterministically encode text into so-called canonical token sequences, to which the LLMs assign probability values. One common assumption is that the probability of a piece of t
Externí odkaz:
http://arxiv.org/abs/2408.08541
Probabilistic Circuits (PCs) are a general framework for tractable deep generative models, which support exact and efficient probabilistic inference on their learned distributions. Recent modeling and training advancements have enabled their applicat
Externí odkaz:
http://arxiv.org/abs/2406.00766
Autor:
Ahmed, Kareem, Teso, Stefano, Morettin, Paolo, Di Liello, Luca, Ardino, Pierfrancesco, Gobbi, Jacopo, Liang, Yitao, Wang, Eric, Chang, Kai-Wei, Passerini, Andrea, Broeck, Guy Van den
Structured output prediction problems are ubiquitous in machine learning. The prominent approach leverages neural networks as powerful feature extractors, otherwise assuming the independence of the outputs. These outputs, however, jointly encode an o
Externí odkaz:
http://arxiv.org/abs/2405.07387
Neuro-symbolic AI bridges the gap between purely symbolic and neural approaches to learning. This often requires maximizing the likelihood of a symbolic constraint w.r.t the neural network's output distribution. Such output distributions are typicall
Externí odkaz:
http://arxiv.org/abs/2312.03905
High-quality labels are often very scarce, whereas unlabeled data with inferred weak labels occurs more naturally. In many cases, these weak labels dictate the frequency of each respective class over a set of instances. In this paper, we develop a un
Externí odkaz:
http://arxiv.org/abs/2311.13718
Autor:
Qian, Chendi, Manolache, Andrei, Ahmed, Kareem, Zeng, Zhe, Broeck, Guy Van den, Niepert, Mathias, Morris, Christopher
Message-passing graph neural networks (MPNNs) emerged as powerful tools for processing graph-structured input. However, they operate on a fixed input graph structure, ignoring potential noise and missing information. Furthermore, their local aggregat
Externí odkaz:
http://arxiv.org/abs/2310.02156
Autor:
Roaa Mikdad Nemat, Ahmed Kareem Joma
Publikováno v:
Zanco Journal of Medical Sciences, Vol 28, Iss 2, Pp 282-291 (2024)
Background and objective: Obesity is a multifactorial disease. It is a multifactorial pathology that can be related to an altered nutritional behaviour or secondary to genetic, hypothalamic, iatrogenic or endocrine diseases. Obesity and visual acuity
Externí odkaz:
https://doaj.org/article/1db582a030a041e5b85fb80f55c29e44
Numerous neuro-symbolic approaches have recently been proposed typically with the goal of adding symbolic knowledge to the output layer of a neural network. Ideally, such losses maximize the probability that the neural network's predictions satisfy t
Externí odkaz:
http://arxiv.org/abs/2302.14207
Autor:
Ahmed kareem, Faris J.Wafar
Publikováno v:
Academic Science Journal, Vol 2, Iss 4 (2024)
The main objective of this paper is to study the dynamic behavior of the hepatitis C virus to control the spread of this disease. The study was conducted by creating a new stochastic mathematical model. Three equilibriums specific to this system are
Externí odkaz:
https://doaj.org/article/663d391d4b0e4b3ba269580c81e3a2cc